给出了一种基于模糊神经网络的模型参考自适应控制方案,首先,构造了一种运用递推预报误差（RPE）算法的多层前向神经网络,并用其对被控对象建模,然后,又构造了一种模糊神经网络控制器（FNNC）,从而为一类难以建立精确数学模型的非线性被控对象提供了一种新的自适应控制方法,仿真结果验证了其有效性, A model reference adaptive control scheme based on fuzzy neural network is proposed. Firstly, a multi-layer neural network with recursive forecasting error (RPE) algorithm is constructed and then used to model the controlled object. Then, , A kind of fuzzy neural network controller (FNNC) is constructed, which provides a new adaptive control method for a class of nonlinear controlled objects which is difficult to establish an accurate mathematical model. The simulation results verify its effectiveness,